1734. Cytokine levels in bronchoalveolar lavage fluid of rhinovirus infected hematopoietic cell transplant recipients: associations with mortality
Session: Oral Abstract Session: Transplant and Immunocompromised Hosts: Emerging Issues
Saturday, October 6, 2018: 9:45 AM
Room: W 2002

Background: HRV lower respiratory tract infection (LRTI) is associated with significant mortality in HCT recipients. The associations between specific cytokine responses and mortality after HRV LRTI are not known.

Methods: Stored frozen BALF samples from adult and pediatric HCT recipients with HRV detected in BALF were analyzed for 30 cytokines (Mesoscale, Rockville, MD). An elasticnet model with Cox regression was used to identify cytokines and other covariates most associated with overall and pulmonary mortality at 90 days, including cytopenias, baseline oxygen (O2) use, copathogens, steroid use, viral load, and viral species. Identified variables were evaluated in multivariable models and the impact of each variable on the bootstrapped optimism corrected concordance statistic (c-statistic) was calculated. Cytokine levels as outcomes were evaluated using multivariable linear regression.

Results: BALF from 84 HCT recipients with HRV detected in BALF from 1998-2015 were included in the analysis. Variables identified in the elasticnet model included: baseline O2, monocyte count, lymphocyte count, steroid use, endothelial neutrophil activator 78 (ENA-78), IL-4, IL-15, IFN- a2a and MCP-2. Viral load and species were not associated with mortality. In the model with the highest c-statistic, baseline O2, monocyte count, lymphocyte count, steroid use and ENA-78 [adjusted hazard ratio (aHR) 1.19, 95% confidence interval (CI) 1.05-1.35] were significantly associated with overall mortality. For pulmonary mortality, the same clinical factors (except for steroids) and IL-4 (aHR 1.27, 95% CI 1.06-1.54) were associated with the outcome (Figure 1). Models including multiple cytokines did not improve the c-statistic. IL-4 levels were associated with viral load but not with host or clinical factors (slope 0.36, 95% CI 0.1-0.61) (Figure 2).

Conclusion: Elevated IL-4 levels in BALF of HCT recipients with HRV LRTI were associated with increased risk of day 90 pulmonary mortality and may provide unique prognostic information. Viral load and species were not associated with mortality, suggesting host immune responses play a larger role than viral factors in disease severity. Cytokines may be possible targets for intervention.

Text Box: Figure 1
Text Box: Figure 2

Alpana Waghmare, MD1,2,3, Isaac Jenkins, MS1, Brad Edmison, BA4, Tillie Loeffelholz, BA4, Terry Stevens-Ayers, MS5, Alex Greninger, MD PhD3, Jane Kuypers, PhD3, Keith Jerome, MD, PhD3,5, Janet Englund, MD, FIDSA6,7, Wendy Leisenring, ScD1 and Michael Boeckh, MD, PhD, FIDSA3,5, (1)Fred Hutchinson Cancer Research Center, Seattle, WA, (2)Seattle Children's Hospital, Seattle, WA, (3)University of Washington, Seattle, WA, (4)Fred Hutchinson Cancer Research Center, Seatte, WA, (5)Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center, Seattle, WA, (6)Pediatrics, Seattle Children's Hospital, Seattle, WA, (7)Department of Pediatrics, University of Washington, Seattle, WA

Disclosures:

A. Waghmare, Ablynx: Investigator , Research support .

I. Jenkins, None

B. Edmison, None

T. Loeffelholz, None

T. Stevens-Ayers, None

A. Greninger, None

J. Kuypers, None

K. Jerome, None

J. Englund, Gilead: Consultant and Investigator , Consulting fee and Research support . Novavax: Investigator , Research support . GlaxoSmithKline: Investigator , Research support . Alios: Investigator , Research support . MedImmune: Investigator , Research support .

W. Leisenring, None

M. Boeckh, Asun Biopharma: Consultant and Investigator , Consulting fee and Research support . Gilead Sciences: Consultant and Investigator , Consulting fee and Research support . Chimerix Inc.: Consultant and Investigator , Consulting fee and Research support . Humabs: Consultant , Consulting fee . GSK: Investigator , Research support .

<< Previous Abstract | Next Abstract

Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.